Improved sleep stage predictions by deep learning of photoplethysmogram and respiration patterns

Kianoosh Kazemi, Arash Abiri, Yongxiao Zhou, Amir Rahmani, Rami N. Khayat, Pasi Liljeberg, Michelle Khine

Abstract: Highlights•Clinical sleep staging requires polysomnography in a sleep lab, which is time-consuming, expensive, uncomfortable, and limited to a single night.•Sensor technology advances have enabled home sleep monitoring but it lacks sufficient accuracy to inform clinical decisions.•We developed a deep learning architecture to accurately classify 5 stages of sleep by combining a convolutional neural network and bidirectional long short-term memory.•Leveraging accessible photoplethysmography signals with respiratory inputs, we significantly improved prediction accuracy and Cohen's kappa for sleep classification.